Professional Documents
Culture Documents
ABSTRACT that are definitely useless for the user this is mainly due to the fact
With the tremendous growth of information available to end users that the very basic relevance criterions underlying their
through the Web, search engines come to play ever a more critical information retrieval strategies rely on the presence of query
role. Nevertheless, because of their general-purpose approach, it is keywords within the returned pages. When a user enters a query
always less uncommon that obtained result sets provide a burden composed by the following keywords “hotel,” “Rome,” and
of useless pages. The next-generation Web architecture, “historical center” (or “hotel,” “Roma,” and “centrostorico”) in
represented by the Semantic Web, provides the layered the Italian version of the well-known Google search engine [7].
architecture possibly allowing overcoming this limitation. The He or she would not be astonished probably by finding that the
ontology for multiple search engines is written such that in this result set actually includes several hotels located in the historical
search engine for single query the final result is got from multiple center of Rome, as expected small town at some distance from the
search engines. After getting the user query result we can use the Rome city center is also included. However, two hotels located in
clustering. In this clustering the user query results is formed in the the historical center of other main Italian cities are also displayed.
a to z form, the several search engines have been proposed, which Finally, three hotels named Roma are included among the 10 most
allow increasing information retrieval accuracy by exploiting a relevant results even if they have nothing to do with the selected
key content of Semantic Web resources, that is, relations. We can city. Only 4 out the 10 results presented to the user satisfy user
use web cache optimization in search engine to get fast retrieval of needs. (Even if they seem to satisfy the user query, based on the
user query results. In this work I have used web cache strategy adopted to process it). Currently, the Semantic Web, (i.e.
optimization based on eviction method for semantic web search online documents written in RDF or OWL), is essentially a web
engine. In this paper, analization of both advantages and universe parallel to the web of HTML documents. Semantic Web
disadvantages of some current Web cache replacement algorithms documents (SWDs) are characterized by semantic annotation and
including lowest relative value algorithm, least weighted usage meaningful references to other SWDs [5]. Since conventional
algorithm and least unified-value algorithm is done. Based on our search engines do not take advantage of these features, a search
analysis, we proposed a new algorithm, called least grade engine customized for SWDs, especially for ontology’s, is needed
replacement (LGR), which takes recency, frequency, perfect- by human users as well as by software agents and services[3]. At
history, and document size into account for Web cache this stage, human users are expected to be semantic web
optimization. researchers and developers who are interested in accessing,
exploring and querying RDF and OWL documents found on the
web.
KEYWORDS:
Semantic web, multiple search engines, ontology, clustering, web
caching, LRU, LGR, LUV algorithm. 2. WEB CACHE DESIGN
We discuss three general purpose cache distributions and lookup
1. INTRODUCTION enhancements that improve both the locality and latency of cache
advertisements. The system uses a form of hierarchical
The Semantic Web is known for being a web of Semantic Web
aggregation to summarize the contents of cached files available in
documents; however, little is known about the structure or growth
a particular local area network. In this way the amount of indexing
of such a web. Search engines such as Google have transformed
the way people access and use the web and have become a critical information that has to be exported to other systems in a WAN
technology for finding and delivering information [1]. Most can be reduced. A common criticism of distributed hash tables is
existing search engines, however, provide poor support to that they lack locality. This is a side effect of the hash function
accessing the web of result’s and make no attempt to take used to identify both nodes and content in the DHT network. The
advantage of the structural and semantic information encoded in hash function provides a key-to node mapping that distributes
SWDs. The Semantic Web will offer the way for solving this keys uniformly at random across the address space of the DHT
problem at the architecture level. In fact, in the Semantic Web, [10]. As such, semantically related nodes and data items when
each page possesses semantic metadata that record additional processed by a systems hash function will be mapped to random
details concerning the Web page itself. This work is designed to locations in the network with high probability. This presents a
serve the research activities in Semantic Web community, number of problems for cache index and lookup systems. First,
especially the following: lookup requests for file content such as images and linked web
pages require a separate lookup request for each URL. This will
(i) Multiple Search Engine for single user query result in a worst-case time complexity of O (M log N) where M is
(ii) Apply Clustering Method the number of embedded file references in a webpage and N is the
(iii) Unwanted pages in the result set would force him or her to number of nodes in the system. Second, due to the random nature
perform a post processing on retrieved information to discard of the hash functions used to identify files, lookup requests for
unneeded ones. Today, search engines constitute the most helpful linked files are likely to be routed to nodes that are far away in the
tools for organizing information and extracting knowledge from network. This significantly adds to the latency of locating a
the Web. However, it is not uncommon that even the most cached file in the network. However, many of these lookup
renowned search engines return result sets including many pages requests are unnecessary and can be reduced by exploiting the link
9
International Journal of Computer Applications (0975 – 8887)
Volume 10– No.9, November 2010
structure of web pages. In a typical web browsing scenario, client Ying, Edward, and Ye-sho argued that model-driven simulation
Software will make a connection to a web server and download was more objective than trace-driven. A web cache algorithm
the HTML specification of a web document. Once this has called Least Weighted Usage (LWU) was proposed using model-
occurred, the client process will parse the document tree and driven simulation [11,12].
generate a series of HTTP get requests to download embedded file 3. Least Unified Value Algorithm (LUV)
content from the web server. As such, this content should also be Bahn et al. proposed a web cache replacement algorithm
available at the same remote cache system as the main webpage called LUV that uses complete reference history of documents, in
unless it has been evicted from the remote cache. To reduce these terms of reference frequency and recency [11,12].
extraneous lookup requests, cache misses and extra round-trip
delays, we have developed a combined indexing structure that Disadvantage of Existing System
client systems can use to identify the set of related cache items (i) Text based searching example (Google, yahoo, msn,
also available at a remote site. This combined index has been Wikipedia).
implemented using a bitmap vector the contents of which are used (ii) Without semantic relationship to give exact result.
to determine the presence or absence of linked web content. This (iii) Query only focus single search engine.
effectively allows a client system to select a remote cache based (iv) Most existing search engines however, provide poor
upon the number of related documents that it stores. As a result, support to accessing the web results.
lookup requests for related files such as embedded images can be (v) No analysis of stopping keywords from the user query.
downloaded from the same remote cache without having to (vi) It will not give relevant or exact result.
specifically locate the file using the DHT index. The idea here is (vii) Number of iterations is high.
to extend the reach of the cache index by one link’s worth, to (viii)A replacement policy is required for replacing a page
enable a client system to determine ahead of time whether linked from web cache to make room for new page.
content is available at a remote proxy. As a consequence,
communication between a client and remote cache system can be 2.2 PROPOSED SYSTEM
reduced because of these cache hints. This allows a client system
The Semantic Web will offer the way for solving this problem at
to maintain a persistent connection with a remote cache, so that
the architecture level. In fact, in the Semantic Web, each page
file requests for linked web content can be pipelined across the
possesses semantic metadata that record additional details
same socket. To create this combined index, the link structure of a
concerning the Web page itself. It will be proved that relations
cached file has to be extracted using regular expressions. This
among concepts embedded into semantic annotations can be
process creates an ordered set of links that can be used to create a
effectively exploited to define a ranking strategy for Semantic
bitmap vector of the linked files available at a remote site. As
Web search engines. A similarity score measuring the distance
such, the length of a bitmap vector corresponds to the number of
between the systematic descriptions of both query and retrieved
out links in a given web page[12]. To encode the availability of
resources is defined. They first explode an initial set of relations
linked content at a remote site, the corresponding bit locations of
(properties) by adding hidden relations, which can be inferred
these out links are set in the bitmap. Therefore, the ith link is
from the query. Similarity is then computed as the ratio between
represented by the ith bit in the bitmap vector. To illustrate this
relation instances linking concepts specified in the user query and
idea, consider a web page that has five links to other files. If each
actual multiplicities of relation instances in the semantic
of these linked files were available at a remote cache, then each
knowledge base. This method is applied on each property
bit location in the bitmap vector of this cached item would be set
individually and requires exploring all the Semantic Web
to one. However, if only the second and third links were available
instances. Moreover, the user is requested to specify all the
at a remote cache, then only bit locations one and two would be
relations of interest. Thus, since it is predictable that the number
set in the bitmap. The intuition here is that users will browse to a
of relations will largely exceed the number of concepts, its
new page through an existing hyperlink directly, instead of
Applicability in real contexts is severely compromised. A similar
jumping to a new page at random[11].Therefore, if we know
approach, aimed at measuring the relevance of a semantic
which links are available ahead of time, the number of cache
association (that is, a path traversing several concepts linked by
lookup messages routed across the network can be reduced. Once
semantic relations) [3]. We provide an interesting definition of
a browser has downloaded a list of IP addresses and adjacency
relevance as the reciprocal of the ambiguity of the association
cache bitmaps from the DHT, these are added to a fixed size in
itself. Ontology-based lexical relations like synonyms, antonyms,
memory cache which has a least recently used eviction strategy.
and homonyms between keywords (but not concepts) have been
used to “expand” query results which automatically associate
2.1 EXISTING SYSTEM related concepts, and exploit the semantic knowledge base to
Nevertheless, because of their general-purpose approach, it is automatically formulate formal queries.
always less uncommon that obtained result sets provide a burden This work web cache (proxy server) is to develop a
of useless pages. It is not uncommon that even the most utility to share internet from single connection to a large network
renowned search engines return result sets including many pages around 200 machines with different operating systems. The
that are definitely useless for the user this is mainly due to the fact software is developed using the Java Language. Java applet
that the very basic relevance criterions underlying their applications are mostly used in the web pages, but we use JFC
information retrieval strategies rely on the presence of query (swing) for developing the software [10].
keywords within the returned pages [10]. This work provides an intelligent environment
1. Lowest Relative Value Algorithm (LRV) containing a number of ready-made options like cache, log file,
Luigi and Vicisano proposed a replace algorithm for error checking, connection pooling, etc. These ready-made tools
proxy cache called Lowest Relative Value (LRV). It is based on may be any of the GUI components that are available in the Java
maximizing an objective function for the whole cache. The AWT package. By using this utility, Administrator can control
objective function uses a cost/benefit model to calculate the and maintain the whole network. This thesis aim is to use the
relative value of each document in the cache. Two performance Least Recently Used document in web caches which replaces
parameters of cache are used: the HR and BHR [11,12]. Randomized web cache replacement algorithm. A web cache sits
2. Least Weighted Usage Algorithm (LWU) between web server and a client and watches request for web
10
International Journal of Computer Applications (0975 – 8887)
Volume 10– No.9, November 2010
pages. It caches web documents for serving previously retrieved 3.1 Web Search Engine Design
pages when it receives a request for them. The term "search engine" is often used generically to
describe both crawler-based search engines and human-powered
2.3 SEMANTIC WEB INFRASTRUCTURE directories. These two types of search engines gather their listings
in radically different ways. Crawler-based search engines, such as
Google, create their listings automatically [2]. They "crawl" or
"spider" the web, then people search through what they have
found. A human-powered directory, such as the Open Directory,
depends on humans for its listings. When we submit a short
description to the directory for your entire site or editors write one
for sites they review. A search looks for matches only in the
descriptions submitted.
3.1.1 Web Crawler
A search engine cannot work without a proper index where
possible searched pages are stored, usually in a compressed
format. This index is created by specialized robots, which crawl
the Web for new/modified pages (the actual crawlers, or
spiders)[4]. Typical crawler architecture is depicted in the figure2
below.
Fig 1: ArchitectureDiagram.
11
International Journal of Computer Applications (0975 – 8887)
Volume 10– No.9, November 2010
citeceer etc., to name a few. At first, the search engines were mentioned before it is more reliable, more advantageous than the
selected and the user query is submitted to all search engines existing one which uses the old Data structures concept. It can
under consideration. The queries covered a broad range of topics. work in a larger network and also maintains load balancing so I
The topics are as follows: Computer science, education, Internet, conclude that this system application is executable under any
literature, music, plants, sports, travel etc. The precision of platform and with any number of clients too.
content of these pages is compared to give the result.
4.1 HTTP CONTENT AND PARSING
3.1.3 Design of Ontology Search ANALYSIS
As mentioned in the last section, finding ontologies to satisfy user Parsing is the process of analyzing an input sequence in order to
requirements is a very important issue, in both KB reuse and determine its grammatical structure with respect to a given formal
Semantic Web areas. There is no existing tool to solve this grammar [7].
problem. Google does have the power, but does not seem to be
specific enough to give good results [5]. After some experiments, PROCESS STEPS
we noticed that the problem arises because Google does not offer
a good visualization function for the ontology files (in different 4.1.1Lexical analysis:
formalisms, such as RDFs, etc.), as the user cannot view the
ontology in an intuitive graphic format; they have to look thro ugh The input character stream is split into meaningful
the ontologies as structured text files. This process takes a lot of symbols (tokens) defined by a grammar of regular
time and cannot guarantee a good result, as the plain text of the expressions. Example: the lexical analyzer takes
ontology cannot show the internal structure of the ontology "12*(3+4)^2" and splits it into the tokens 12, *, (, 3, +, 4, ), ^
clearly. The ontology searching steps is shown in the figure 3 and 2.
below.
4.1.2 Syntax analysis
It performs checking if the tokens form an legal
expression, with respect to a CF grammar. Limitations are it
cannot check (in a programming language): types or proper
declaration of identifiers
12
International Journal of Computer Applications (0975 – 8887)
Volume 10– No.9, November 2010
engines; such as Google, Yahoo, AltaVista, Clusty, Excite, All the 5. FETCH UNITS AND RESULT PRE-
web, File tube, You tube, Amazon, Cite seer, Wikipedia, Isohunt
etc., here the answer is got from 25 search engines, searches take FETCHING
place simultaneously from different search engines, the main In many search engine architectures, the computations required
advantage is that searches do not have to wait for each search during query execution are not greatly affected by the number of
engine, the results are computed simultaneously from multiple results that are to be prepared, as long as that number is relatively
search engines and they are displayed on the screen. The result is small. In particular, it may be that for typical queries, the work
therefore much faster. required to fetch several dozen results is just marginally larger
than the work required for fetching 10 results. Since fetching more
results than requested may be relatively cheap, the dilemma is
whether storing the extra results in the cache (at the expense of
evicting previously stored results) is worthwhile. Roughly
speaking, result prefetching is profitable if, with high enough
probability, those results will be requested shortly - while they are
still cached and before the evicted results are requested again. One
aspect of result prefetching was analyzed in, where the
computations required for query executions (and not cache hit
ratios) were optimized [10].
The Figure 6 shows the web cache main GUI. Here java
swing is used to build the user interface. It has four menus, Admin
menu, View menu, Setup menu and Help menu.
The Admin menu is shown in figure. It has four
submenus. They are Start server to start HTTP server, Stop server
to stop the HTTP server, Clear cache clears manually all the cache
files in your cache directory, Clear image manually clears all the
images in your image directory.
The View menu has four submenus, Server response to
get the server log Information and also monitoring the connection
between server and internet you can save log file for future
reference, Client request monitors the connection between the
local server and number of clients. This process monitors every
client request response and also identifies which ask the web
URL’s and also save log information for future reference. The
The comparison table shows search engine performance for the picture viewer is used to view all the pictures in your cached
user query what is java swing? That user query gets the result picture directory. The window is used to see each single image
from different search engines. I can measure the user query based size and dimension and also to delete the unwanted single image
on normal query result, without any semantic analysis and web from the cached image directory.
cache optimization analysis. Then the same query get different Cache view is used to view the cache file from the cached
result from different search engine using web cache optimization. directory and then view file and save the cached file in different
Then the same user query get the different search engine result formats, for example html, jsp and asp.
based on semantic web result. Then I can form the bar chart using The Setup menu gives information about the server
three column values. In a similar way a line chart is formed for the configuration which has two Submenus, The network input
same query with three column values to analyze the performance. configuration window were we can set the proxy input and port
number and the connection properties can be set like receive
13
International Journal of Computer Applications (0975 – 8887)
Volume 10– No.9, November 2010
timeout, send time out number log of threads, cached url’s and Proc. 14th Int’l Conf. World Wide Web (WWW ’05), pp.
cache expires (hrs) time. It is used for refreshing the server every 117-127, 2005.
one hour using N-M algorithm in LRU method.The logging
window gives the access log and error log information. The access [3] T. Berners-Lee, J. Hendler, and O. Lassila, “The Semantic
log caches the errors based on pages got from URL. The error log Web,” scientific Am., 2001.
shows connection error between client and server. The logging [4] S. Brin and L. Page, “The Anatomy of a Large-Scale
level is based on minimal normal and detail information about the Hypertextual Web Search Engine,” Proc. Seventh Int’l Conf.
error. The Help menu shows the information about the server. World Wide Web (WWW ’98), pp. 107-117, 1998.
[5] Seda Cakiroglu, Erdal Arikan, “Replace Problem in Web
6. CONCLUSION Caching”, in Proceedings of IEEE Symposium on Computers
The next-generation Web architecture represented by the and Communications, June, 2003.
Semantic Web will provide adequate instruments for improving
search strategies and enhance the probability of seeing the user [6] S. Cohen, J. Mamou, Y. Kanza, and Y. Sagiv, “XSEarch: A
query satisfied without requiring tiresome manual refinement. Semantic Search Engine for XML,” Proc. 29th Int’l Conf.
Nevertheless, they mainly use page relevance criteria based on Very Large Data Bases, pp. 45-56, 2003.
information that has to be derived from the whole knowledge
[7] Berners-lee, t., Hendler, j., and lassila, o. (2001) "The
base, making their application often unfeasible in huge semantic
Semantic Web". Scientific American, May
environments. By neglecting the contribution of the remaining
annotated resources, a reduction in the cost of the query answering http://www.sciam.com/2001/0501issue/0501berners-lee.html
phase could be expected. Despite the promising results in terms of [8] L. Ding, T. Finin, A. Joshi, R. Pan, R.S. Cost, Y. Peng, P.
both time complexity and accuracy, further efforts will be Reddivari, V. Doshi, and J. Sachs, “Swoogle: A Search and
requested to foster scalability into future Semantic Web Metadata Engine for the Semantic Web,” Proc. 13th ACM
repositories based on multiple ontology, characterized by billions Int’l Conf. Information and Knowledge Management (CIKM
of pages, and possibly altered through next generation “semantic” ’04), pp. 652-659, 2004.
spam techniques. It has been designed and partially implemented
to capture more metadata on classes and properties and to support [9] L. Ding, T. Finin, A. Joshi, Y. Peng, R. Pan, and P.
millions of documents. We have also built an ontology dictionary Reddivari, “Search on the Semantic Web,” Computer, vol.
based on the ontologies discovered by our research, which we 38, no. 10, pp. 62-69, Oct. 2005.
continue to refine. We have described a prototype crawler-based [10] H. Bahn, S. Noh, S. L. Min, and K. Koh, “Using Full
indexing and retrieval system for Semantic Web documents. The Reference History for Efficient Document Replacement in
traditional browsing cache systems can not address both non Web Caches”, in Proceedings of the 2nd USENIX
stationary and stationary browsing behaviors at the same time. Symposium on Internet Technologies & Systems, October,
The response time for an interactive browsing system can be 1999.
greatly increased.
[11] Ying Shi, Edward Watson, and Ye-sho Chen, “Model-
7. REFERENCES Driven Simulation of World-Wide-Web Cache Policies”, In
Proceeding of the 1997 Winter Simulation Conference, June,
[1] B.Aleman-Meza, C. Halaschek, I. Arpinar, and A. Sheth,
1997.
“A Context-Aware Semantic Association Ranking,” Proc.
First Int’l Workshop Semantic Web and Databases (SWDB [12] Ganesh, Santhanakrishnan, Ahmed, Amer, Panos K.
’03), pp. 33-50, 2003. Chrysanthis and Dan Li, “GDGhOST: A Goal Oriented Self
[2] K. Anyanwu, A. Maduko, and A. Sheth, “SemRank: Ranking Tuning Caching Algorithm”, in Proceeding of the 19th ACM
Complex Relation Search Results on the Semantic Web,” Symposium on AppliedComputing.March, 2005.
14